Taxonomical nomenclature historically relies on morphological recognition. The number of nominally recognized taxa per group may depend on the number of available traits (morphological complexity) to compare in the group, accordingly, without additional information of distribution, ecology and/or genetics. This artifact hypothesis remains controversial. In the case of species, their morphological differences and thus the number of available traits in their genera may reflect the time length since speciation, if they are geographically or reproductively isolated. Instead, by using the numbers of subspecies per species, we could minimize similar confounding effects of history in testing the predicted dependence on the number of available traits. In land snails, for example, the mean number of subspecies (excluding nominotypical subspecies) per species is 0.16 in the Zonitidae and 4.0 in the Chondrinidae. We tested hypotheses of the artifact and of the other factors that may underlie variation in the mean number of subspecies among six European families.